The available data strongly suggest that the causes of psychiatric illnesses are complex, and that the risk of suffering from depression, schizophrenia, anxiety disorder and other psychiatric diseases is influenced by genetic inheritance, nongenetic biological factors and external environmental factors such as social stress. In addition, it is clear that monoamine neurotransmitters (serotonin, dopamine and norepinephrine) are related to the onset and treatment of depression, anxiety disorders and other psychopathologies. Despite the evidence for genetic influences on psychiatric disorders, on levels of monoamine neurotransmitters, and on normal variation in temperament related to disease, the specific genes that affect these traits are not well known. Whole genome scanning using linkage analysis in multi-generation pedigrees is a powerful method for locating functional genes that influence complex traits such as these. Unfortunately, for several reasons, this approach cannot be used with human families to locate genes that influence cerebrospinal fluid (CSF) levels of monoamines, or to investigate normal variation in behavior. In this project, we propose to conduct a whole genome linkage scan in a nonhuman primate model (baboons, Papio hamadryas). We will search for genes that influence CSF levels of monoamine metabolites (5-HIAA, HVA and MHPG) and also investigate individual variation in temperament by subjecting each baboon to a behavioral challenge involving response to novel objects. All 650 study animals have already been genotyped for a linkage map consisting of 350 human microsatellite loci, with 7 cM resolution. We will also use gene expression array methods to assess the molecular effects of identified QTL loci in prefrontal cortex. Preliminary results from about 300 baboons indicate that all three monoamine metabolites and several behavioral responses to challenge are strongly heritable. Most significantly, the available genotypes permit a preliminary genome scan, and four LOD scores greater than 1.9 have been obtained, including a LOD score of 2.4 for the dopamine metabolite HVA, and 2.6 for an anxiety-related behavioral trait. Our preliminary data demonstrate the value of the baboon model and indicate that a larger sample from the same pedigrees would likely provide important new information about genes that influence both monoamine neurotransmitter levels and behavioral reactivity (i.e., temperament). Identification of these genes will be very significant for future studies of genetic risk factors for psychiatric illness in humans.